1. ONLINE SHOPPING SYSTEM In an online shopping svetem they are using to attract customers based on adyertismonts, If more customers will shen they get more profit, to attract more customers they are...

Extracted text: 1. ONLINE SHOPPING SYSTEM In an online shopping svetem they are using to attract customers based on adyertismonts, If more customers will shen they get more profit, to attract more customers they are using many techniques like special discounts for regular customers, product recommendations based on previous purchases and fast and reliable delivery opportunities etc. Data mining techniques are used in an online shopping process to make suggestions such as 'who bought this also bought these', by using the other customer's purchases. In this scenario, when the customer wants to buy a product, he will be able to see the products that previous customers bought together with that product. Thus, this customer will be more likely to purchase one of these products. In this way, the target audience will be reashed and product sales will be increased. Visitors' number of visits to the online shop and number of purchased items is calculated for each visitor, and thus a label can be created whether the visiteris a buyer or not. Thus, when a new visitor arrives, it can be understood whether she is a buyer or not by looking the visiters, with same qualification. Here the input is a number of visiters, with same qualification and the output is to classify based on their previous visits. d) What are some possible and meaningful reasons for at least some of the atributes in your database having missing values? e) What are some reasons for noisy data, outliers, and data inconsistencies? f) How do you handle these problems in the data cleaning phase? (miccing value handling, noisy data handling, resolving inconsistencies and detecting outiers) 9) Are there any intearation, problems? (shema, integration, data value conflicts, redundant attributes) Do you need to apply anv-data reduction strategies? h) Do you define new variables, not existing in the original database, to selve the classification (regression) problems? (foature creation) ) Are there some features carrying similar informatien?Are tbrere irrelevant features? ) Do you apply any samling methods? k) What data transformations and/or dieeritization0s are needed? ) Evaluate the possible solution of one of the data mining problems from the business point of view. ) Suppose one of these, three problems is solved successfully. Describe the deployment of the solution in the real environment. What are some possible impacts of the data mining solution? Can you imagine any unanticipated situations after implementation of the solution?